Hello!
Our favourite links this month include:
Also, CEA has made major updates to effectivealtruism.org — you can leave feedback here. Registration is now open for EAGxSāoPaulo (22–24 August) and EA Summit: Santiago (19 July). And, as always, we’re sharing jobs, opportunities, great articles, videos and podcasts.
P.S.: We’ve almost hit 60,000 subscribers. If you value this newsletter, click the link below to forward it to a friend and help us get there.
|
|
|
— Toby, for the EA Newsletter Team
|
|
Articles
Why AGI might still be a long way off
This newsletter often highlights forecasts suggesting that AGI could arrive surprisingly soon. But Zershaaneh Qureshi’s latest piece for 80,000 Hours offers three reasons that it might still take decades.
First, today’s AI systems can’t manage long, complicated tasks the way humans can. They don’t adapt easily or learn from their experience. Dwarkesh Patel points to the lack of “continual learning” — AI’s inability to build up context and improve over time — as a key reason why today’s systems can’t fully replace human workers. Perhaps these issues are unavoidable under our current approach to AI development, not just today’s generation of agents. In that case, it could take time to figure out how to get further.
Second, the pace of progress might slow down. Recent gains have relied on steady supplies of compute, data, and investment — all of which could stall.
Third, the world doesn’t seem to be preparing for AGI. Markets aren’t pricing in massive disruption, and governments aren’t moving quickly to reshape institutions. Even among AI experts, there is no consensus about AGI arriving soon.
All that said, the longer timelines to AGI are much shorter than they were a few years ago. Even sceptical experts predict AGI within the next few decades, whereas we used to debate whether it would arrive this century.
|
|
Crunch time for cage-free
Over the past decade, more than 2,700 companies — including Walmart, Kroger, and McDonald’s — pledged to stop selling eggs from caged hens by the end of 2025. Lewis Bollard and Emma Buckland ask: How’s it going?
Many major companies are backing off and making excuses. Target cites high prices. Kroger blames slow customer uptake. Others, like Subway and Best Western, haven’t released data in years. Several US states have delayed their bans on caged eggs. The European Commission missed its deadline to propose an EU-wide ban. The UK’s Labour government remains noncommittal.
And yet... the overall trajectory remains positive. Over the past decade, the share of cage-free hens has risen from 13% to 45% in the US, from 44% to 62% in the EU, and from 50% to 82% in the UK — sparing over 300 million hens from cages. Most companies with pledges have followed through, including McDonald’s, Starbucks, Amazon, and all five of the US’s largest food manufacturers.
Advocates now aim to hold laggards accountable, continue global progress and cement gains in law.
Want to act now?
|
|
Progress on lead continues
In this newsletter last October, I wrote about lead poisoning. It remains a significant problem that affects one in three children worldwide, but we are making progress.
Paint, spices, cosmetics, cookware, batteries — many common products in low- and middle-income countries still contain lead, decades after it was banned elsewhere. The result is slowly devastating: reduced IQ, higher rates of cardiovascular disease, and millions of premature deaths. In some hotspots, blood lead levels are 30 times higher than in the US.
The good news is that this is not a mysterious public health problem. As ‘The end of lead’ argues, we already know what to do — screen products for lead, regulate, substitute, repeat. Rich countries (mostly) have done it. Poorer ones haven’t, largely for lack of data, capacity, and technical support.
LEEP (the Lead Exposure Elimination Project) has been focusing on lead paint. In 2024, they expanded to 31 countries and helped pass new regulations in Nigeria and Liberia. LEEP’s programmes in 13 countries are estimated to avert 9.6 million DALYs (disability-adjusted life years) at a cost of $4.49 each. A new initiative, LABRI, has been launched this year to address the same issue for battery recycling — another major source of exposure, particularly for children living near informal smelting sites.
Lead is a fixable problem. It just hasn’t been fixed yet.
Read LEEP’s 2024 review, or the long-form case in Works in Progress.
|
|
In other news
For more stories, try these email newsletters and podcasts.
|
|
Resources
Links we share every time — they're just that good!
|
|
Jobs
Boards and resources:
Selection of jobs
80,000 Hours
ALLFED
Centre for Effective Altruism
EA UK
- Director (Remote (London preferred), GBP £45K–£62.5K, apply by June 27th)
Fish Welfare Initiative
Givewell
Mirror Biology Dialogues Fund
- Director of Policy (Remote (Washington, DC preferred), USD $150K–200K, apply by June 29th)
Open Philanthropy
|
|
Announcements
Bonus: Do you listen to (or watch) The 80,000 Hours Podcast? Would you like to join its advisory group and give feedback on new episodes? The group influences who the podcast team interview, what topics they cover, and how they talk about them. Subscribe to receive surveys after each episode comes out.
Events and Conferences
Fellowships and Courses
- Apply for the Effective Thesis Fellowships 2025 to develop a high-impact thesis with expert mentorship. Programmes are online and paid, and start September–October 2025.
- Longview Philanthropy, Macroscopic Ventures, and The Navigation Fund are offering the Research Fellowships on Digital Sentience for PhD- or JD-level researchers to explore AI consciousness and related legal or social questions. The paid programme runs online or in-person for up to two years.
Prizes and Funding
- Career Transition Fellowships on Digital Sentience offer funding, resources, and networking for professionals or students pivoting into digital mind ethics. The fellowships are paid and remote-friendly — the deadline to apply is July 9th.
- Researchers and practitioners are invited to apply for funding for applied projects on AI consciousness, moral status and digital sentience. This paid, remote opportunity is open to individuals or orgs; apply by July 9th.
|
|
Organizational Updates
You can see updates from a wide range of organizations on the EA Forum.
|
|
Timeless Classic
We’re highlighting the progress of cage-free campaigns this month, so it’s a great time to share this fantastic podcast episode from 2021 with Leah Garcés, on her experience working with the chicken industry to institute change. Notably, Leah tells the story of her fruitful friendship with a chicken farmer, someone she initially thought was her natural enemy.
|
|
We hope you found this edition useful!
If you’ve taken action because of the Newsletter and haven’t taken our impact survey, please do — it helps us improve future editions.
Finally, if you have any feedback for us, positive or negative, let us know!
Errata: Last month I wrote that the word for the practice of eating insects was 'Entophagy'. In fact, it is 'Entomophagy'. I apologise for any awkward moments I caused.
– Toby, for the EA Newsletter Team |
|
|
|
|
